This is a competing continuation application to further HIV prevention on the Internet as a risk environment. Based on our findings in Men's INTernet Study (MINTS-I) we identified the following next steps: Aim I: To investigate the generalizability and stability of Internet-mediated risk behavior in MISM, we propose a cross-sectional, block design study of 3,750 MISM to assess how HIV risk variables interact in Internet-mediated liaisons. Analysis testing differences across age, race/ethnicity, HIV serostatus, and rural-urban residence is proposed. Aim II: To develop a theoretically sound, highly interactive, state-of-the-art, Internet-based HIV prevention intervention for MISM, we propose to partner with a leading software development company to adapt a theoretically-sound, scientifically validated seminar-based HIV prevention intervention into a high quality, Internet-based one. We propose an eight stage SAVVY design and development process using our Usability Laboratory facilities (N=60 MISM). Aim III: To test the effects of the new, Internet-based HIV prevention intervention for MISM on knowledge, attitudes and unsafe sexual behavior, we propose a randomized controlled trial (N=250 per condition) with 3, 6, and 12-month follow-up surveys. Strengths of this proposal include a highly productive team experienced in intervention development and Internet-sex research partnering with leaders in e-learning, a superb environment in which to study Internet-risk, a highly innovative project (including the development and testing of one of the first Internet-based HIV prevention interventions), and the potential of the new intervention to lower HIV risk globally among English-reading MISM.